Problem
Research questions and friction points this paper is trying to address.
online handwriting generation
handwriting style
stroke continuity
character connectivity
sentence-level synthesis
CASHG: Context-Aware Stylized Online Handwriting Generation
🤖 AI Summary
Generating sentence-level online handwriting trajectories that are stylistically consistent, contextually coherent, and stroke-smooth remains challenging. This work proposes CASHG, the first method to explicitly model inter-character connectivity and spacing by integrating a character context encoder, a bi-character-aware sliding-window Transformer decoder, and a gated context fusion mechanism to achieve high-quality generation. The approach employs a three-stage curriculum training strategy and introduces CSM, a novel boundary-aware metric for evaluating connectivity and spacing. Experimental results demonstrate that CASHG significantly outperforms existing methods on the CSM metric while maintaining competitive performance in DTW-based trajectory similarity, with human evaluations further confirming its superior generation quality.
📝 Abstract
Online handwriting represents strokes as time-ordered trajectories, which makes handwritten content easier to transform and reuse in a wide range of applications. However, generating natural sentence-level online handwriting that faithfully reflects a writer's style remains challenging, since sentence synthesis demands context-dependent characters with stroke continuity and spacing. Prior methods treat these boundary properties as implicit outcomes of sequence modeling, which becomes unreliable at the sentence scale and under limited compositional diversity. We propose CASHG, a context-aware stylized online handwriting generator that explicitly models inter-character connectivity for style-consistent sentence-level trajectory synthesis. CASHG uses a Character Context Encoder to obtain character identity and sentence-dependent context memory and fuses them in a bigram-aware sliding-window Transformer decoder that emphasizes local predecessor--current transitions, complemented by gated context fusion for sentence-level context.Training proceeds through a three-stage curriculum from isolated glyphs to full sentences, improving robustness under sparse transition coverage. We further introduce Connectivity and Spacing Metrics (CSM), a boundary-aware evaluation suite that quantifies cursive connectivity and spacing similarity. Under benchmark-matched evaluation protocols, CASHG consistently improves CSM over comparison methods while remaining competitive in DTW-based trajectory similarity, with gains corroborated by a human evaluation.
Innovation
Methods, ideas, or system contributions that make the work stand out.
context-aware handwriting generation
inter-character connectivity
sliding-window Transformer
curriculum learning
connectivity and spacing metrics
🔎 Similar Papers
No similar papers found.
